Cancer immunotherapy using adoptive T cell transfer has shown its promises in the treatment of hematological malignancies. Despite tremendous efforts, results of clinical trials using tumor infiltrating lymphocytes or engineered CAR-T cells in human solid tumors have not produced desired therapeutic responses. In solid tumors, there are multiple barriers that prevent cytotoxic T cells reaching tumor cells. First, abnormal blood vessels limit T cell delivery into solid tumors, especially tumor center. The presence of dense tumor stromal cells and extracellular matrix create a physical barrier that traps T cells in an immunosuppressive stroma microenvironment. Therefore, there is an unmet need to develop novel approaches to improve delivery efficiency of cytotoxic T cells in solid tumors. The translational goal of the proposed phase I research is to develop and validate a stroma-breaking therapeutic T cell delivery system for immunotherapy in human solid tumors with a focus on pancreatic and colon cancers. We have developed a receptor targeted and stroma breaking drug delivery platform with novel designs to overcome tumor stromal cellular and extracellular matrix barriers. We have demonstrated its ability to deliver nanoparticle-drug crossing tumor vessels, migrating through tumor stroma, and penetrating the basement membrane to reach tumor cells. This recombinant targeting ligand consists of urokinase plasminogen activator receptor (uPAR) binding domain of uPA fused with the catalytic domain of matrix metalloproteinase-14 (ATFmmp14). Our preliminary results showed that co-delivery of T cells with ATFmmp14 ligands led to significant increases in the numbers of T cells delivered in tumors following intravenous delivery in mouse colon and pancreatic tumor models as well as a human colon cancer patient derived xenograft (PDX) model. In this phase I project, we aim to determine the effect of ATFmmp14-T cell delivery system on intratumoral delivery and distribution using the MHC-non-restricted cytotoxic ?? T cells as a T cell model system in human colon and pancreatic cancer PDX models. First, we will develop the best formulation of ATFmmp14 conjugated T cell delivery platform using ?? T cells obtained from healthy donors in human pancreatic and colon PDX models (Aim1). Next, we will determine whether ATFmmp14 ligand can improve selective delivery of ?? T cells into pancreatic and colon PDX tumors and promote cell migration through stroma and basement membrane barriers to reach tumor cells (Aim 2). Biodistribution and systemic toxicity in normal organs and tissues will also be evaluated (Aim 2). Finally, the therapeutic efficacy of targeted delivery of ?? T cells without or with the combination of a chemotherapy drug will be determined in pancreatic and colon PDX models (Aim 3). Results of the above proposed studies should allow us building a strong foundation for conducting further translational studies in the Phase II research project to bring this approach into a clinical trial. Result of this study will also lead to future development and commercialization of this unique approach for other adaptive T transfer therapies, such as CAR-T, tumor specific T cells, and tumor infiltrating T cells.